Income Polarization and Asking the Right Question

An article on Yahoo News today addressed the truth to Obama's statements that the rich have been getting richer while the poor get poorer (Economic Study). The article is based on Richard Burkhauser's work. One of his articles that has already been published is similar to the Yahoo article. The one specifically referenced in Yahoo, however, is still coming down the pike.

The article that is already available is "Presidential Address: Evaluating the Questions That Alternative Policy Success Measures Answer" (a bit of an unwieldy title, huh?). Burkhauser makes an important point that at the bottom of every answer is a question, and how one asks the question often greatly affects the answer. So the question "are the rich getting richer while the poor get poorer?" is a great question for casual conversation, but when it comes to actually answering it, things get much more complicated. Specifically, how should we measure how rich or poor someone is? Usually we just look at income, but as Burkhauser points out, this can often be too simplistic.

Burkhaused has an issue with the convenient use of the "bottom 90% and top 10%" as definitions of rich and poor. He prefers to use quintiles (5 groups separated by 20%). Sounds good. Another problem Burkhauser has is with using just income as a judge. He says:

However, this measure of

income growth does not recognize that tax units are a subset of households and that

income sharing can occur across tax units within households. (For example, two

unmarried persons who live together must file their income tax forms separately.

Doing so makes them two separate sharing units in the Piketty and Saez world of

column 1, but they are nevertheless likely to share their market income between

them in their single household.)

Again, he may be onto something. He also believes we should include "in-cash government transfers." Taking these three new additions to the measure of rich or poor:

When we then acknowledge that households are of different sizes and that the

income available to a given individual will be affected by the number of persons in

his or her household (returns to scale are not perfect) and adjust our measure of

income in column 4 accordingly, we find that the increase of the median person’s

household size–adjusted pre-tax, post-government in-cash income increased by 23.6

percent, more than seven times the growth in column 1. The definition of income

used in column 4 is the one most often used in the United States poverty, income,

and income inequality literatures.

OK, I can dig it. Maybe those sounding the alarm about income polarization are a bit hyperbolic. Burkhauser then goes on to add fringe benefits to the mix. This makes income polarization even less likely.

So in addressing what the proper question should be, Burkhauser has revised the conventional approach and added some important ways to better judge income polarization. But has he arrived yet at the right question? No where does Burkhauser address hours worked per week. If the lower 20% are working twice as many hours a week for an increase in his revised income measure of 26.4%, that isn't a real increase. Or if a middle income earner spent 3 years going to school to get a degree and spent $60k of their own money to increase their human capital, that too must be included in the question. Other examples would be easy to come up with.

I appreciate Burkhauser's casuistry, but two things must go along with his revisions. One is the realization that his question still hasn't gotten to the bottom of the Question. And secondly, the more details and specifics that are included (and even those which are left out) can make it much more difficult to tell if a researcher is cherry-picking data in order to arrive at a pre-arranged solution. I am in no way implying that is what Burkhauser has done, simply that it needs addressed.